SAR Automatic Target Recognition Using Transfer Learning Approach

Mufti, Maha Al; Hadhrami, Esra Al; Taha, Bilal; Werghi, Naoufel

doi:10.1109/icoias.2018.8494149

Cited by 18 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Xu et al [39] proposed a differentiated adaptive regularized transfer learning framework for SAR ship classification to overcome the limitation under insufficient labeled training samples. Al Mufti et al [40] employed a pretrained AlexNet to train a multiclass SVM classifier.…”

Section: Transfer Learning Transfer Learningmentioning

confidence: 99%

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Zhai

Deng

et al. 2019

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Though Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) via Convolutional Neural Networks (CNNs) has made huge progress toward deep learning, some key issues still remain unsolved due to the lack of sufficient samples and robust model. In this paper, we proposed an efficient transferred Max-Slice CNN (MS-CNN) with L2-Regularization for SAR ATR, which could enrich the features and recognize the targets with superior performance. Firstly, the data amplification method is presented to reduce the computational time and enrich the raw features of SAR targets. Secondly, the proposed MS-CNN framework with L2-Regularization is trained to extract robust features, in which the L2-Regularization is incorporated to avoid the overfitting phenomenon and further optimizing our proposed model. Thirdly, transfer learning is introduced to enhance the feature representation and discrimination, which could boost the performance and robustness of the proposed model on small samples. Finally, various activation functions and dropout strategies are evaluated for further improving recognition performance. Extensive experiments demonstrated that our proposed method could not only outperform other state-of-the-art methods on the public and extended MSTAR dataset but also obtain good performance on the random small datasets.

show abstract

Section: Transfer Learning Transfer Learningmentioning

confidence: 99%

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Zhai

Deng

et al. 2019

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Wagner et al [18] proposed a network combining a CNN and a SVM, and additional training methods to incorporate prior knowledge. Maha Al Mufti et al [19] used a pretrained AlexNet and a SVM as the classifier for SAR ATR. Furukawa et al [20] utilized the deep residual network (ResNet) for SAR ATR.…”

Section: Introductionmentioning

confidence: 99%

A Lightweight Fully Convolutional Neural Network for SAR Automatic Target Recognition

Zhou

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Automatic target recognition (ATR) in synthetic aperture radar (SAR) images has been widely used in civilian and military fields. Traditional model-based methods and template matching methods do not work well under extended operating conditions (EOCs), such as depression angle variant, configuration variant, and noise corruption. To improve the recognition performance, methods based on convolutional neural networks (CNN) have been introduced to solve such problems and have shown outstanding performance. However, most of these methods rely on continuously increasing the width and depth of networks. This adds a large number of parameters and computational overhead, which is not conducive to deployment on edge devices. To solve these problems, a novel lightweight fully convolutional neural network based on Channel-Attention mechanism, Channel-Shuffle mechanism, and Inverted-Residual block, namely the ASIR-Net, is proposed in this paper. Specifically, we deploy Inverted-Residual blocks to extract features in high-dimensional space with fewer parameters and design a Channel-Attention mechanism to distribute different weights to different channels. Then, in order to increase the exchange of information between channels, we introduce the Channel-Shuffle mechanism into the Inverted-Residual block. Finally, to alleviate the matter of the scarcity of SAR images and strengthen the generalization performance of the network, four approaches of data augmentation are proposed. The effect and generalization performance of the proposed ASIR-Net have been proved by a lot of experiments under both SOC and EOCs on the MSTAR dataset. The experimental results indicate that ASIR-Net achieves higher recognition accuracy rates under both SOC and EOCs, which is better than the existing excellent ATR methods.

show abstract

“…In addition, the Automatic Target Recognition (ATR) is a considerable interest in many industrial applications. Some constraints, such as real time response, high detection accuracy, adaptive capabilities to noisy environments, increase the complexity of the target detection task (Al Mufti et al, 2018). These methods can be used in many applications and investigated based on the conditions of the problem.…”

Section: Introductionmentioning

confidence: 99%

An Intelligent Location and State Reorganization of Traffic Signal

Behzadi

2020

Geodesy and Cartography

View full text Add to dashboard Cite

In all geo-database related to traffic, beside storing roads data, the information associated to traffic signals such as location, types of traffic signals, side street name, and so on are also stored in that database. In reality, the reason of defining traffic signals for road is the situations and conditions which the roads have. So the existence of traffic signals in the network is related to the parameters of the road. In this paper, instead of storing traffic signal data in the database, a novel method is introduced which implemented on the road network. As a result, the spatial and non-spatial information of traffic signals in the network are extracted based on the location and attribute of the road network. The proposed method is implemented on the network; the result of the intelligent method is compared with the traffic signals information which stored in the database. By comparing the locations and states of proposed traffic signals and the real ones, the overall accuracy for recognizing locations of traffic signal is obtained 94% and the overall accuracy for recognizing states of traffic signal is obtained 89%.

show abstract

SAR Automatic Target Recognition Using Transfer Learning Approach

Cited by 18 publications

References 14 publications

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

A Lightweight Fully Convolutional Neural Network for SAR Automatic Target Recognition

An Intelligent Location and State Reorganization of Traffic Signal

Contact Info

Product

Resources

About

SAR Automatic Target Recognition Using Transfer Learning Approach

Cited by 18 publications

References 14 publications

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L2-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L2-Regularization

A Lightweight Fully Convolutional Neural Network for SAR Automatic Target Recognition

An Intelligent Location and State Reorganization of Traffic Signal

Contact Info

Product

Resources

About

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization

Robust SAR Automatic Target Recognition Based on Transferred MS-CNN with L²-Regularization